考虑决策者偏好信息的配电网多目标无功优化研究  

Research on Multi-Objective Reactive Power Optimization of Distribution Network Considering Decision Maker's Preference

作　　者：高菲[1] 宋晓辉[1] 盛万兴[1] 孟晓丽[1] 

机构地区：[1]中国电力科学研究院,北京市海淀区100192

出　　处：《电网技术》2016年第9期2845-2851,共7页Power System Technology

基　　金：国家自然科学基金资助项目(5177152);国家电网公司科技项目(PD71-15-043)~~

摘　　要：为充分利用有限的计算资源提高多目标无功优化的实用性和计算效率,提出一种考虑决策者偏好信息的配电网无功多目标优化方法。该方法将优化过程分为2个阶段,第1阶段完成Pareto前沿的粗略描绘,并利用等值线原理和决策者对目标权值及数值的偏好,形成搜索偏好区域。第2阶段利用偏好区域特性提出改进的支配关系,引导粒子群在偏好区域内进行集中广泛搜索,完成多目标无功优化。基于IEEE33节点系统的算例测试验证了所提方法能够将决策者偏好有效转化为对应的搜索区域,改进的支配关系则保证了计算资源逐步收缩集中在该搜索区域内。与传统多目标无功优化方法相比,对世代距离、分布性、错误率以及迭代次数等各项指标均有较大幅度提升。Inorder to make full use of limited computing resourcesto improve engineeringpracticability and efficiency, this paper proposes amethod ofmulti-objective reactive power optimizationconsidering preference information of decision maker.This method is divided into two stages. In the first stage, approximate Pareto solution setisobtained withlimited iterations. Then preferredsearching area is constructed, based oncontour lineprincipleanddecisionmaker＆#39;spreferencein objective weight and value. In the second stage,optimization focuswithin thepreferredsearching area is applied based on improved dominance relation.The case in IEEE 33-node system verifies thatdecision maker＆#39;s preference is converted to corresponding searchingarea, and theimproved dominance relationguarantees that the computing resources are gradually shrinking in thepreferredsearchingarea.Compared with traditional multi-objective reactive power optimization, the method proposed in this paper has a large improvement in generation distance, spacing, error rateand reduces optimization time.

关 键 词：多目标 无功优化 偏好信息 等值线 改进支配关系 

分 类 号：TM721[电气工程—电力系统及自动化]